[PATCH] net/vhost: get csum offload capabilities of vhost backend

[Hu, Jiayu]

Hi Wenwu,

> -----Original Message-----
> From: Ma, WenwuX <wenwux.ma at intel.com>
> Sent: Thursday, February 17, 2022 11:16 PM
> To: maxime.coquelin at redhat.com; Xia, Chenbo <chenbo.xia at intel.com>
> Cc: dev at dpdk.org; Hu, Jiayu <jiayu.hu at intel.com>; Wang, Yinan
> <yinan.wang at intel.com>; He, Xingguang <xingguang.he at intel.com>; Ma,
> WenwuX <wenwux.ma at intel.com>
> Subject: [PATCH] net/vhost: get csum offload capabilities of vhost backend
> 
> The current vhost backend lacks csum offloads information, which will cause
> testpmd command such as "csum set tcp hw <port_id>" to fail. This patch
> adds the information according to the device features.

Vhost-pmd cannot report correct checksum offloading capabilities, since
feature negotiation may happen after vhost-pmd device creation. It's good
to see there is a solution for it, IMO.

> 
> Signed-off-by: Wenwu Ma <wenwux.ma at intel.com>
> ---
>  drivers/net/vhost/rte_eth_vhost.c | 18 ++++++++++++++++++
>  1 file changed, 18 insertions(+)
> 
> diff --git a/drivers/net/vhost/rte_eth_vhost.c
> b/drivers/net/vhost/rte_eth_vhost.c
> index 070f0e6dfd..7593d5a9ae 100644
> --- a/drivers/net/vhost/rte_eth_vhost.c
> +++ b/drivers/net/vhost/rte_eth_vhost.c
> @@ -1281,6 +1281,24 @@ eth_dev_info(struct rte_eth_dev *dev,
>  				RTE_ETH_TX_OFFLOAD_VLAN_INSERT;
>  	dev_info->rx_offload_capa = RTE_ETH_RX_OFFLOAD_VLAN_STRIP;
> 
> +	if (internal->vid != -1) {
> +		uint64_t features = 0;
> +		if (rte_vhost_get_negotiated_features(internal->vid,
> &features) != 0)
> +			return 0;
> +
> +		if (features & (1ULL << VIRTIO_NET_F_CSUM)) {
> +			dev_info->tx_offload_capa |=
> RTE_ETH_TX_OFFLOAD_TCP_CKSUM |

The spec says "VIRTIO_NET_F_CSUM: host handles pkts w/ partial csum".
If this feature is negotiated, it means vhost pmd supports to receive packets
with partial checksum, right? So VIRTIO_NET_F_CSUM should represent Rx
offloading capabilities for vhost pmd.

> +
> 	RTE_ETH_TX_OFFLOAD_UDP_CKSUM |
> +
> 	RTE_ETH_TX_OFFLOAD_IPV4_CKSUM;
> +		}
> +
> +		if (features & (1ULL << VIRTIO_NET_F_GUEST_CSUM)) {
> +			dev_info->rx_offload_capa |=

Same as above.

Thanks,
Jiayu